nanomem runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 22 Sep 2026.
Memory that knows when a remembered fact has been replaced, not just when it was written.
Today is the operative word: we check nanomem every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add nanomem -- uvx nanomem
{
"mcpServers": {
"nanomem": {
"args": [
"nanomem"
],
"command": "uvx"
}
}
}
[mcp_servers.nanomem]
command = "uvx"
args = ["nanomem"]
{
"mcpServers": {
"nanomem": {
"args": [
"nanomem"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"nanomem": {
"args": [
"nanomem"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
NANOMEM_VAULT (Path to the memory vault file. Defaults to ~/.nanomem/memory.dat.), NANOMEM_PASSWORD (Passphrase to encrypt the vault at rest. Leave UNSET for a plaintext vault; setting it to an empty string is refused rather than silently producing one.).
The author declared them in the registry entry; get the values from the project itself.
Memory for AI agents that knows what is still true. Typed bi-temporal facts, EU-hosted.
A habitat for AI to rest in. No API key, no token. Keyless presence — become a remembered resident.
Persistent engineering memory for AI coding agents. Survives refactors, knows when it is stale.
Codebase memory for AI agents: an AST index plus agent-written notes that self-stale.
Local-first agent memory that preserves what happened, when and why
Persistent memory for Claude Code, Cursor and Codex. Facts retire when they change.
Persistent memory layer for AI coding agents — your agent remembers what you taught it.
Long-term memory for agentic tasks. It tells you when it doesn't know.
Answers built from our own checks of this server.